The name of the game in healthcare management is this: Do more with less.
Take antibiotics, for example. Here's a stiff managerial challenge: Put vastly more patients on the costly infection-fighting drugs, but spend less and use fewer total doses.
At the same time, decrease the death rate and the incidence of adverse drug events associated with antibiotics.
Do all this while holding the line on that alarming new threat to clinical effectiveness: bacterial resistance to commonly used antibiotics.
Hospitals around the country are writing these kinds of objectives into their game plans for trimming wasted effort from their medical routines, usually by combining clinical protocols with computerization.
Will the effort pay off? LDS Hospital in Salt Lake City is reporting that the objectives listed above not only can be met but have been met during the course of a seven-year initiative.
Using clinical and financial information systems to feed data constantly to physicians, the hospital over time achieved near perfection in timing the use of antibiotics to prevent infections during surgery, according to a study published last month in the Annals of Internal Medicine.
That precision timing resulted in fewer post-operative infections, which the hospital converted into quick cutoff of standing antibiotic orders and a drop in the average number of doses per patient to 5.3 in 1994 from 19 in 1988.
In addition, the merging of computerized patient data with antibiotic information helped pinpoint which antibiotics to use in treatment, when to use them and at what dosage, said Stanley Pestotnik, clinical pharmacist at LDS' department of clinical epidemiology.
During the study period, the proportion of patients receiving antibiotics rose each year, to 53% in 1994 from 32% in 1988. That was a consequence of increasing severity of illness at the 520-bed teaching hospital, Pestotnik said. LDS' Medicare case-mix index jumped to 2.05 in 1993 from 1.75 in 1988 (See chart).
Yet the total use of antibiotics, measured in defined daily doses, fell nearly 23%. And antibiotic costs per treated patient, when adjusted for inflation, fell to $52 in 1994 from $123 in 1988, according to the study.
During that period, the death rate related to antibiotics decreased to 2.65% in 1994 from 3.65% in 1988. Adverse drug events associated with antibiotics fell 30%.
Reaping savings. Measured just in financial terms, the progress made in curbing spending on antibiotics went straight to the bottom line.
Hospitals are major users of antibiotics to prevent or combat infections, from examination rooms to surgery suites and inpatient areas.
Those antibiotics can dominate the prescription drug budget, accounting for 20% to 50% of hospital spending on drugs, according to recent industry figures cited in the LDS study. The sicker the patients, the higher the likelihood they'll need costly antibiotics, Pestotnik said.
At LDS, spending on antibiotics reached $988,000 in 1988, about 25% of the $4 million racked up for all pharmacy drug expenses.
By 1994, drug spending had risen to $7.2 million, or $4.8 million when adjusted for inflation.
But antibiotic expenditures fell slightly during that span to $925,000 in 1994. When adjusted for inflation, antibiotic spending actually plummeted to $612,000 in 1988 dollars, and it represented only 13% of total drug spending by 1994.
Information as catalyst. The pivotal forces behind these results revolved around using compelling and credible information to help staff physicians come up with their own clinical solutions rather than try to enforce or control medical decisions, Pestotnik said.
There were no restrictive formularies of drugs. Clinical practice guidelines written into the information systems were agreed upon and periodically revised by the physicians themselves, using a continuous loop of new medical knowledge generated by local practice as well as published findings, he said.
The hospital developed four programs to support clinical decisions on use of antibiotics:
For surgical use, to prevent wound infections.
For suspected infection without lab data, to predict what microbes may be lurking and suggest treatments.
For antibiotic therapy based on confirmed bacterial infection.
For antibiotic safety, taking dosages and interactions into account when ordering.
The programs then computed the most cost-effective regimen of therapy for each patient and fed the choices back to the physician.
Because the hospital's information system kept track of all patients as far back as 1985, the study was drawn from all 162,000 patients discharged between Jan. 1, 1988, and Dec. 31, 1994, of which nearly 64,000 patients received antibiotics.
The pervasiveness of the computer system was its central feature.
"These programs continually track and assist physicians in managing each patient treated with an antibiotic at LDS Hospital and in all aspects of antibiotic use," the study said. "No antibiotic can be prescribed at LDS Hospital without being affected by these decision-support programs."
Discoveries. When it was first implemented, the decision-support system was generating an average of 2.67 alerts per day, and physicians were changing therapy 30% of the time based on the information provided.
Pestotnik said the relatively low batting average wasn't an indicator of physician resistance but rather was traceable to a high proportion of "false positives" generated by the system's screening software. That's because initial screens were based on national practice guidelines, which often didn't apply to the patients and medical environment of the hospital, he said.
Once the guidelines were retooled based on local practices and information supplied by the LDS computer database, the average number of alerts fell to 1.32 by 1994, and physicians changed their therapy based on the alerts in 99.9% of cases, the study said.
Another improvement in antibiotic use came in surgery, where a number of medical studies have shown that patients are at risk for wound infections if they don't get preventive antibiotics within two hours of the opening incision.
LDS was meeting that mark in only 40% of surgery cases in 1985, when its surgical decision-support system was implemented, Pestotnik said. By 1994, the posted percentage was 99.1%.
After surgery, physicians write standing orders for a follow-up series of antibiotic doses, he said. If patients aren't infected, there's no need to keep giving them the drugs. But the hospital discovered early on that physicians would forget to cancel the order.
Now the computer system monitors patient post-operative conditions and alerts hospital pharmacists to track down surgeons on their morning rounds and remind them to stop the antibiotics, Pestotnik said.
The net result has been a reduction in the total "tonnage" of antibiotics used in the hospital environment, he said. And LDS clinicians involved in the study said that's one reason they've been able to stabilize the incidence of antibiotic-resistant pathogens during the seven-year span at a time when medical literature was reporting the trend as an increasing problem.
Reducing overall antibiotic use gives pathogens less of a chance to develop resistant strains, Pestotnik said. And by encouraging choice and feedback rather than narrowing choices for cost reasons, the hospital avoids expanding the use of a limited number of options.
"Restrictive policies on drugs have actually increased the resistance pressures on antibiotics," he said.